Accelerating Fractional PINNs using Operational Matrices of Derivative
This work addresses computational bottlenecks in fractional differential equation modeling for researchers in scientific computing, though it is incremental as it builds on existing fPINN methods.
The paper tackles the slow training of fractional Physics-Informed Neural Networks (fPINNs) by introducing an operational matrix method that replaces automatic differentiation with a precomputed matrix-vector product, achieving enhanced accuracy, particularly with the Legendre Neural Block architecture, across various differential equations like DDEs and DAEs.
This paper presents a novel operational matrix method to accelerate the training of fractional Physics-Informed Neural Networks (fPINNs). Our approach involves a non-uniform discretization of the fractional Caputo operator, facilitating swift computation of fractional derivatives within Caputo-type fractional differential problems with $0<α<1$. In this methodology, the operational matrix is precomputed, and during the training phase, automatic differentiation is replaced with a matrix-vector product. While our methodology is compatible with any network, we particularly highlight its successful implementation in PINNs, emphasizing the enhanced accuracy achieved when utilizing the Legendre Neural Block (LNB) architecture. LNB incorporates Legendre polynomials into the PINN structure, providing a significant boost in accuracy. The effectiveness of our proposed method is validated across diverse differential equations, including Delay Differential Equations (DDEs) and Systems of Differential Algebraic Equations (DAEs). To demonstrate its versatility, we extend the application of the method to systems of differential equations, specifically addressing nonlinear Pantograph fractional-order DDEs/DAEs. The results are supported by a comprehensive analysis of numerical outcomes.